Automated multi-layer optical design via deep reinforcement learning

Wang, Haozhu; Zheng, Zeyu; Ji, Chengang; Guo, L. Jay

doi:10.1088/2632-2153/abc327

Cited by 51 publications

(26 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Additional model development beyond these studied models is necessary to incorporate fabrication robustness as a learning objective. For example, by re-parametrizing the structures 57 , or building suitable datasets and incorporating the fabrication variation into loss functions 58 , it is possible for neural networks to learn these properties and output predicted structures that are robust to fabrication variations.…”

Section: Discussionmentioning

confidence: 99%

Benchmarking deep learning-based models on nanophotonic inverse design problems

Ma¹,

Tobah²,

Wang³

et al. 2022

OES

Self Cite

View full text Add to dashboard Cite

Photonic inverse design concerns the problem of finding photonic structures with target optical properties. However, traditional methods based on optimization algorithms are time-consuming and computationally expensive. Recently, deep learning-based approaches have been developed to tackle the problem of inverse design efficiently. Although most of these neural network models have demonstrated high accuracy in different inverse design problems, no previous study has examined the potential effects under given constraints in nanomanufacturing. Additionally, the relative strength of different deep learning-based inverse design approaches has not been fully investigated. Here, we benchmark three commonly used deep learning models in inverse design: Tandem networks, Variational Auto-Encoders, and Generative Adversarial Networks. We provide detailed comparisons in terms of their accuracy, diversity, and robustness. We find that tandem networks and Variational Auto-Encoders give the best accuracy, while Generative Adversarial Networks lead to the most diverse predictions. Our findings could serve as a guideline for researchers to select the model that can best suit their design criteria and fabrication considerations. In addition, our code and data are publicly available, which could be used for future inverse design model development and benchmarking.

show abstract

Section: Discussionmentioning

confidence: 99%

Benchmarking deep learning-based models on nanophotonic inverse design problems

Ma¹,

Tobah²,

Wang³

et al. 2022

OES

Self Cite

View full text Add to dashboard Cite

show abstract

“…For completeness we want to mention also work on reinforcement learning for iterative design optimization, where the neural network learns to behave as an iterative optimization algorithm. The expectation is that the ANN can adapt its optimization strategy specifically to the given problem and hence outperform conventional heuristic algorithms [77,78].…”

Section: B Direct Neural Network Inverse Designmentioning

confidence: 99%

Deep learning in nano-photonics: inverse design and beyond

et al. 2021

View full text Add to dashboard Cite

Deep learning in the context of nano-photonics is mostly discussed in terms of its potential for inverse design of photonic devices or nanostructures. Many of the recent works on machine-learning inverse design are highly specific, and the drawbacks of the respective approaches are often not immediately clear. In this review we want therefore to provide a critical review on the capabilities of deep learning for inverse design and the progress which has been made so far. We classify the different deep learning-based inverse design approaches at a higher level as well as by the context of their respective applications and critically discuss their strengths and weaknesses. While a significant part of the community's attention lies on nano-photonic inverse design, deep learning has evolved as a tool for a large variety of applications. The second part of the review will focus therefore on machine learning research in nano-photonics "beyond inverse design". This spans from physics informed neural networks for tremendous acceleration of photonics simulations, over sparse data reconstruction, imaging and "knowledge discovery" to experimental applications. http://dx.doi.org/XX.XXXX/XX.XX.XXXXXX CONTENTS C Deep learning for interpretation of photonics experiments 11 4 Conclusions and perspectives 13

show abstract

“…Expanding on this idea, very recently Wankerl et al made significant improvements to the optimization of multi-layered thin films by introducing a multi-path deep Q-learning algorithm to handle both discrete (material types, number of layers) and continuous (layer thicknesses) parameters, and these parameters were also recently considered by Wang et al using a deep RL sequence generation network via an RNN variant. This allows for the consideration of the entire physical structure together which eliminates the need to reduce the input parameter space [133,134]. Thus, RL has been proven to be a useful algorithm for solving inverse design problems, and inverse design is highly prevalent in the design of plasmonic devices.…”

Section: Perspectives On Future Workmentioning

confidence: 99%

Instantaneous Property Prediction and Inverse Design of Plasmonic Nanostructures Using Machine Learning: Current Applications and Future Directions

Aggarwal

Shankar

2022

Nanomaterials

View full text Add to dashboard Cite

Advances in plasmonic materials and devices have given rise to a variety of applications in photocatalysis, microscopy, nanophotonics, and metastructures. With the advent of computing power and artificial neural networks, the characterization and design process of plasmonic nanostructures can be significantly accelerated using machine learning as opposed to conventional FDTD simulations. The machine learning (ML) based methods can not only perform with high accuracy and return optical spectra and optimal design parameters, but also maintain a stable high computing efficiency without being affected by the structural complexity. This work reviews the prominent ML methods involved in forward simulation and inverse design of plasmonic nanomaterials, such as Convolutional Neural Networks, Generative Adversarial Networks, Genetic Algorithms and Encoder–Decoder Networks. Moreover, we acknowledge the current limitations of ML methods in the context of plasmonics and provide perspectives on future research directions.

show abstract

Automated multi-layer optical design via deep reinforcement learning

Cited by 51 publications

References 24 publications

Benchmarking deep learning-based models on nanophotonic inverse design problems

Benchmarking deep learning-based models on nanophotonic inverse design problems

Deep learning in nano-photonics: inverse design and beyond

Instantaneous Property Prediction and Inverse Design of Plasmonic Nanostructures Using Machine Learning: Current Applications and Future Directions

Contact Info

Product

Resources

About